Exploring Emotion Contagion with “Breathing” Robots
Introduction
People often cling to others when they’re scared. For example, a child might clutch a parent after being startled, whereas adults may grab a partner’s arm during a tense scene in a horror movie. This isn’t just a random habit—holding onto someone can help calm us down when we are frightened (e.g., Coan et al. 2006). However, in some situations, I argue that physical touch may fail to downregulate negative emotional experiences—specifically, when an individual being touched is expressing fear themselves.
We built plush robots with motorized ribcages that expanded and contracted to simulate breathing. These robots exhibited three types of breathing patterns:
Stable Breathing: The robot’s breathing rate was maintained at approximately 14 breaths per minute, simulating a calm state.
Fearful Breathing: The robot's breathing rate increased gradually from 14 to 30 breaths per minute, simulating hyperventilation associated with fear, before gradually returning to 14 breaths per minute.
No Breathing: The robot remained completely still, serving as a control condition.
Participants held these robots while watching a 90-second fear-inducing video clip from the movie "The Shining". Throughout the experiment, participants’ heart rates (HR) were monitored to determine if the robot’s breathing patterns affected their physiological and emotional responses to the fear stimulus.
Results
Heart Rate Changes: Participants holding the robot exhibiting a fearful breathing pattern demonstrated a significant increase in heart rate compared to those holding the stable-breathing or non-moving robots. Conversely, participants holding the robot exhibiting a stable and calm breathing pattern demonstrated a significantly decreased heart rate when compared to the non-moving control condition. These results suggest that robots exhibiting a rapid and fearful breathing pattern can heighten humans’ fear, whereas a robot demonstrating a stable and calm breathing pattern can reduce humans’ fear. Human physiological responses are therefore guided by the robot’s behavior.
Discussion
Real-World Applications
These findings have important implications for multiple real-world scenarios:
Human-Robot Interaction: Robots can modulate human emotions, highlighting their potential use in therapy, education, and entertainment.
Wearable Technology: Integrating simulated breathing movements into wearable devices could enhance desired fear-inducing or thrilling experiences, such as when viewing horror films or gaming.
Conclusion
Humans can indeed “catch” fear from dynamic robots through touch, which guides how we respond to the world around us. As technology continues to advance, the potential for robots to influence human emotions in both positive and negative ways will become increasingly important. Understanding these dynamics is crucial for developing robots that can effectively support and enhance the human experience.
Researchers:
Zak Witkower (lead), University of Amsterdam
Laura Cang, University of British Columbia
Paul Bucci, University of British Columbia
Karon MacLean, University of British Columbia
Jessica Tracy, University of British Columbia